info.debatty.spark.knngraphs.example.Search Maven / Gradle / Ivy

Go to download
/*
 * The MIT License
 *
 * Copyright 2015 Thibault Debatty.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a copy
 * of this software and associated documentation files (the "Software"), to deal
 * in the Software without restriction, including without limitation the rights
 * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
 * copies of the Software, and to permit persons to whom the Software is
 * furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice shall be included in
 * all copies or substantial portions of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 * THE SOFTWARE.
 */
package info.debatty.spark.knngraphs.example;

import info.debatty.java.graphs.NeighborList;
import info.debatty.java.graphs.Node;
import info.debatty.java.graphs.SimilarityInterface;
import info.debatty.java.stringsimilarity.JaroWinkler;
import info.debatty.spark.knngraphs.ApproximateSearch;
import info.debatty.spark.knngraphs.ExhaustiveSearch;
import info.debatty.spark.knngraphs.builder.Brute;
import info.debatty.spark.knngraphs.builder.DistributedGraphBuilder;
import java.io.IOException;
import java.util.ArrayList;
import java.util.List;
import java.util.Random;
import org.apache.spark.SparkConf;
import org.apache.spark.api.java.JavaPairRDD;
import org.apache.spark.api.java.JavaRDD;
import org.apache.spark.api.java.JavaSparkContext;

/**
 *
 * @author Thibault Debatty
 */
public class Search {

    /**
     * @param args the command line arguments
     * @throws java.io.IOException
     */
    public static void main(String[] args) throws IOException, Exception {
        if (args.length != 1) {
            System.out.println(
                    "Usage: spark-submit --class " +
                    Search.class.getCanonicalName() + " " +
                    "");
        }
        
        String file =  args[0];

        // Graph building parameters
        int k = 10;

        // Partitioning parameters
        int partitioning_iterations = 5;
        int partitioning_medoids = 4;

        // Search parameters
        final int search_k = 10;
        final int max_similarities = 400;

        int search_queries = 10;

        // Similarity measure
        final SimilarityInterface similarity = 
                new SimilarityInterface() {

            public double similarity(String value1, String value2) {
                JaroWinkler jw = new JaroWinkler();
                return jw.similarity(value1, value2);
            }
        };

        // Read the dataset file
        ArrayList strings = DistributedGraphBuilder.readFile(file);

        // Configure spark instance
        SparkConf conf = new SparkConf();
        conf.setAppName("SparkTest");
        conf.setIfMissing("spark.master", "local[*]");
        JavaSparkContext sc = new JavaSparkContext(conf);
        
        // Convert to nodes
        List> dataset = new ArrayList>();
        for (String s : strings) {
            dataset.add(new Node(String.valueOf(dataset.size()), s));
        }

        // Split the dataset between training and validation
        Random rand = new Random();
        ArrayList> validation_dataset = new ArrayList>(search_queries);
        for (int i = 0; i < search_queries; i++) {
            validation_dataset.add(dataset.remove(rand.nextInt(dataset.size())));
        }

        // Parallelize the dataset and force execution
        JavaRDD> nodes = sc.parallelize(dataset);
        nodes.cache();
        nodes.first();

        // Compute the graph (and force execution)...
        DistributedGraphBuilder builder = new Brute();
        builder.setK(k);
        builder.setSimilarity(similarity);
        JavaPairRDD, NeighborList> graph = builder.computeGraph(nodes);
        graph.cache();
        graph.first();

        // Prepare the graph for approximate graph based search
        // (and force execution)
        ApproximateSearch approximate_search_algorithm = 
                new ApproximateSearch(
                        graph, 
                        partitioning_iterations, 
                        partitioning_medoids, 
                        similarity);

        // Prepare exhaustive search
        ExhaustiveSearch exhaustive_search = 
                new ExhaustiveSearch(graph, similarity);
        graph.cache();
        graph.first();


        // Perform some search...
        for (final Node query : validation_dataset) {
            System.out.println("Search query: " + query.value);

            // Using distributed graph based NN-search
            NeighborList neighborlist_graph = 
                    approximate_search_algorithm.search(
                            query, 
                            search_k, 
                            max_similarities);
            System.out.println(
                    "Using graph: " + neighborlist_graph.element().node.value);

            // Using distributed exhaustive search
            NeighborList neighborlist_exhaustive = 
                    exhaustive_search.search(query, search_k);
            System.out.println(
                    "Using exhaustive search: " + 
                            neighborlist_exhaustive.element().node.value);
        }
    }
}